Table 3.

A statistical analyses of exon/intron sizes of a random sample of 100 human non-linear splicing events versus a random sample of 100 exons from all human exons, for which there is no evidence of non-linear splicing

	Exons with evidence for non-linear splicing	Random exons with no evidence of non-linear splicing	P-value
A	20 628 (± 3053)	3159 (± 523)	1.5 × 10−7
B	12 556 (± 2839)	3159 (± 523)	0.002
C	10 067 (± 1973)	3191 (± 644)	0.001
D	15 135 (± 2158)	3191 (± 644)	5.4 × 10−7
E	320 (± 58)	153 (± 16)	0.007
F	280 (± 49)	153 (± 16)	0.017

Numbers shown are the Mean length ± the standard error of the mean (n = 100). An independent samples t-test was used to calculate the P-values for the average sizes between these two samples with a 95% confidence interval. A: comparison of the 5′ intron of random exons, to the 5′ introns of the 3′ exon involved in a non-linear splicing event; B: comparison of the 5′ intron of random exons, to the 5′ introns of the 5′ exon involved in a non-linear splicing event; C: comparison of the 3′ intron of random exons, to the 3′ introns of the 3′ exon involved in a non-linear splicing event; D: comparison of the 3′ intron of random exons, to the 3′ introns of the 5′ exon involved in a non-linear splicing event; E: comparison of random exons, to the 5′ exon involved in a non-linear splicing event; F: comparison of random exons, to the 3′ exon involved in a non-linear splicing event.